A Cortically Inspired Architecture for Modular Perceptual AI

TL;DR

This paper proposes a biologically inspired modular architecture for perceptual AI that enhances interpretability, generalization, and robustness by decomposing perception into specialized interacting modules based on neuroscientific principles.

Contribution

It introduces a cortically inspired blueprint for modular perceptual AI, emphasizing hierarchical predictive feedback and shared latent spaces for improved transparency and reasoning.

Findings

Modular decomposition leads to more stable representations.

The architecture supports explicit inference processes.

Empirical evidence shows improved interpretability.

Abstract

This paper bridges neuroscience and artificial intelligence to propose a cortically inspired blueprint for modular perceptual AI. While current monolithic models such as GPT-4V achieve impressive performance, they often struggle to explicitly support interpretability, compositional generalization, and adaptive robustness - hallmarks of human cognition. Drawing on neuroscientific models of cortical modularity, predictive processing, and cross-modal integration, we advocate decomposing perception into specialized, interacting modules. This architecture supports structured, human-inspired reasoning by making internal inference processes explicit through hierarchical predictive feedback loops and shared latent spaces. Our proof-of-concept study provides empirical evidence that modular decomposition yields more stable and inspectable representations. By grounding AI design in biologically validated principles, we move toward systems that not only perform well, but also support more transparent and human-aligned inference.